Looking at the galaxy group mass range (1E13.75 to 1E14.25 solar masses) I find that:

The star mass fractions in TNG300-1 are around 1-1.5%, however other simulations (e.g. Magneticum) have star mass fractions around 2.5-4%.

Consequently the gas fractions in TNG300-1 are larger, around 12-13%, whereas Magneticum has gas fraction around 6-11%

Can it be that my star / gas mass fraction calculation from the TNG300-1 group catalog shown above is wrong, or do you think this calculation is correct?

From observations I found references pointing to higher gas fractions than the numbers I obtain from the TNG300-1 group catalog, for example see figure 3 of:

I would suggest to look at the various stellar mass/fraction diagnostics in Pillepich+2018 (e.g. Fig 11) and perhaps re-make one or two of them, to check that your calculations are correct. Note that any measurement like this is entirely dependent on, and very sensitive to, the definition (e.g. aperture) chosen, as explored a bit in that paper.

Dear Illustris-TNG colleagues,

I have computed the gas and star mass fractions of the groups in the FoF Halos catalog corresponding to TNG300-1 snapshot 99 as follows:

gasFraction = GroupMass/GroupMassType[0]

starFraction = GroupMass/GroupMassType[4]

Looking at the galaxy group mass range (1E13.75 to 1E14.25 solar masses) I find that:

Can it be that my star / gas mass fraction calculation from the TNG300-1 group catalog shown above is wrong, or do you think this calculation is correct?

From observations I found references pointing to higher gas fractions than the numbers I obtain from the TNG300-1 group catalog, for example see figure 3 of:

https://iopscience.iop.org/article/10.1088/0004-637X/778/1/14

Do you think this can be re-conciliated?

Thank you for your help and kind regards

I would suggest to look at the various stellar mass/fraction diagnostics in Pillepich+2018 (e.g. Fig 11) and perhaps re-make one or two of them, to check that your calculations are correct. Note that any measurement like this is entirely dependent on, and very sensitive to, the definition (e.g. aperture) chosen, as explored a bit in that paper.